System Setup
Hardware
- ABIT AN7, NVIDIA nForce2 Ultra 400, Socket A (11/11/03 BIOS)
- ASUS K8V Deluxe, Socket 754, VIA K8T800, Athlon 64, 1002.007 BIOS
- ABIT IC7, Socket 478, Intel i875, Pentium 4, A18 BIOS
- Intel Pentium 4 3.2GHz, 512KB L2, 16 x 200MHz
- AMD Athlon XP3200+, 512KB L2, 11 x 200MHz
- AMD Athlon 64 Model 3200+, 1MB L2, 11 x 200MHz
- NVIDIA GeForceFX 5900 Ultra
- ATI Radeon 9800XT (412/730)
- Corsair XMS3200LLPT, 2 x 256MB, 2-2-2-6
Software
- Windows XP Professional w/SP1 and current Windows Update patches (as of 17/11/03)
- NVIDIA Detonator FX 52.16 WHQL (ABIT AN7)
- ATI CATALYST 3.9 (ASUS K8V, ABIT IC7)
- VIA Hyperion 4.48v chipset drivers (ASUS K8V)
- NVIDIA nForce2 3.13 Platform Driver (ABIT AN7)
- Intel 5.02.1002 chipset drivers (ABIT IC7)
- DirectX 9.0a Runtime
- HEXUS Pifast v41
- Simplisoft HDTach 2.61
- Kribi Bench 1.19
- Sciencemark 2.0
- 3DMark 2001SE v330
- Quake3 v1.30 HQ (four demo)
- Serious Sam 2 (OpenGL HQ, SDC demo)
- LAME 3.92MMX MP3 Encoding (192CBR, U2's Pop album)
- Realstorm Ray Tracing
- Comanche 4 Demo
- X2: The Threat - Rolling Demo
Due to (slowly) migrating to Radeon 9800XT, the AN7 gets an NVIDIA GeForceFX 5900 Ultra in its AGP slot while a new Radeon 9800XT makes its way to me. The Intel and Athlon 64 comparison platforms get Radeon 9800XT however, a board being present throughout their testing. I'll comment on any difference as needed, but being a system review, the graphics card used has (usually) minimal impact on results.
No Socket A comparison systems, I instead decided to use a different angle with AN7. After benchmarking the board, it showed minimal performance difference to recent Socket A boards like ASUS A7N8X 2.0 Deluxe and the DFI LanParty NFII Ultra B I reviewed recently. So instead of graphs with no real discernable difference between the comparison systems, AN7 gets to fight Model 3200+ Athlon 64 and Pentium 4 3.2C instead. It's a fast nForce2 Ultra 400 board as far as they go, so it'll be nice to see if AN7 can uphold 32-bit Athlon XP performance against its other platform rivals, as we evaluate performance.
So, a mix of comparison systems but something that's still easy to follow, I'll keep you informed throughout. As always, tests were run three times and the median result used for reporting.
